InternVL2-26B: benchmark results
Provider: Shanghai AI Lab. Released 2024-07-04. Access: Open.
Aggregate Rating 1476 ± 19, rank #2715 of 10916 rated models, from 552 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MME - Position | 176.7 | Score | 99.1 |
| OpenVLM AI2D - Eclipses | 100 | Accuracy (%) | 98.6 |
| OpenVLM MMT-Bench - Weapon Recognition | 100 | Score (%) | 98.1 |
| OpenVLM MMBench V1.1 CN - Identity Reasoning | 100 | Accuracy (%) | 97.9 |
| OpenVLM MMT-Bench - Set-of-Mark Recognition | 100 | Score (%) | 97.8 |
| OpenVLM MMBench V1.0 EN - AR | 87.8 | Accuracy (%) | 97.4 |
| OpenVLM CCBench - Food and Clothes | 79.1 | Accuracy (%) | 96.7 |
| OpenVLM TextVQA Val | 82.5 | VQA Score (%) | 96.6 |
| OpenVLM MMT-Bench - Human Interaction Understanding | 90 | Score (%) | 95.9 |
| OpenVLM MMT-Bench - Human Object Interaction Recognition | 90 | Score (%) | 95.6 |
| OpenVLM MMBench V1.1 CN - Action Recognition | 93.2 | Accuracy (%) | 95.4 |
| OpenVLM CCBench - Sketch Reasoning | 93.3 | Accuracy (%) | 95.2 |
Interactive version: theaggregate.ai/model?slug=internvl2-26b · How It Works · Data refreshed daily, snapshot 2026-11-08.